A data-driven geospatial workflow to map species distributions for conservation assessments

Aim Species distribution maps are essential for assessing extinction risk and guiding conservation efforts. However, most come sourced as expert-drawn range maps with known issues of accuracy or are developed with overly complex modelling procedures. Thus, data-driven alternatives that are accessible and reliable are a welcome addition to the spatial conservation toolkit. Here, we developed a geospatial workflow to refine the distribution of a species from its extent of occurrence (EOO) to area of habitat (AOH) within the species range map. The range maps are produced with an inverse distance weighted (IDW) interpolation procedure using presence and absence points derived from primary biodiversity data. Location The Americas (North, South, Central America and the Caribbean). Methods As a case study, we mapped the distribution of 723 resident forest birds in the Americas and assessed their performance in comparison with expert-drawn range maps. We evaluated differences in accuracy, spatial overlap, range map size and derived AOH. Results The geospatial workflow generated IDW range maps with a higher overall accuracy (87% versus 62%) and fewer errors of omission (<1%) and commission (14%) than the expert range maps (28% both errors). The spatial overlap between both datasets was low (35%), but the agreement increased in areas of high probability of occurrence (68%). We did not find significant differences in range size, but the AOH derived from the expert-drawn range maps was consistently smaller than the estimates from the IDW range maps. Main Conclusions Our geospatial workflow provides a straightforward approach to accurately map species ranges and the estimation of area of habitat (AOH) for conservation planning and decision-making. Conversely, procedures that refine expert-drawn range maps to obtain AOH risk producing biased estimates for local-scale applications.